Described in a brief manner the idea driving this project was to tweak and bend the textual data present in the dataset of the NMA’s collection to the point where it becomes something else entirely and that -something else- would be the material from which images will be generated.
The resulting images in this project can be separated into two groups. Group a contains images in which the text is readable and group b images in which the text is so highly condensed that it is rendered unreadable and in which it is just functioning as form. Even though the two groups emerge from the same premise they each explore different ideas and possibilities about the textual data that they employ.
In the images that belong to group a what is explored are the intricate relationships that exist within the NMA’s collection. This exploration is not shown in an explicit manner but rather as suggested paths of connectedness open for interpretation on the part of the viewer. I believe that this openness for interpretation also sheds light on the evocative and poetic possibilities of these texts, which have an entirely different function (one that is solely of an informative nature) within the context of the NMA collection. By recontextualising the texts in the setting of the images they became highly expressive while at the same time maintaining their explanatory character. Some of the relationships explored through the images were material type and object type and collection belonging and object type.
In the images that belong to group b what is explored is free form making through the use of text as material. What was of most concern in these images were compositional elements such as colour, typography type and size. In some images the text was drawn with a pre-established composition in mind, that is to say that it was coded (Pocessing) so the the text would be drawn at certain positions and in others the text was being drawn as the mouse moved on the screen. The latter made for a more fluid and free drawing experience. In this group even though the text is rendered unreadable I did keep a log of what text was being used in each of the compositions, finding this information (content) an important element of the images. **
things learned
The NMA’s collection is comprised of 49230 items, these items belong to 1003 different collections and within these items there is a diversity of 196 materials.
The above information might seem as too tip of the iceberg yet I believe it is a perfect exemplar on the accessibility that digital practices have brought about on the keepings of cultural institutions. While it is virtually impossible for museums to show their collections entirely at once in physical form in a digital form it is thinkable. This project and its brief were informative on the contemporary practice of digital collecting through projects such as Stamen’s Artscope for sfmoma and the library hack project.
Code wise there was a lot to learn in this project through the introduction of quite a few new concepts and simply by the size of the dataset provided. There were the xml and json formats which enable to break the data into smaller chunks. These formats aided at getting specific information one wanted from the dataset. ArrayLists(), HashMap and sort() were also introduced, these helped at sorting and accessing the data in a direct manner. In the case of my project these concepts were very useful in Processing to access the text that I wanted to work with in the images.
strength
One of the strengths of the project is that from early on in the process both on a conceptual and formal level I knew what I wanted to do. This enabled me to concentrate on the images themselves while also giving me the freedom to explore the dataset at will in the search for what text to use or what relationships to present in order to create meaningful images. At times this exploration gave the must unique results in terms of the items I came upon, either because of their unique objet quality, their detailed description or their unusual material. Finding this objects at random while generating the images at times felt like diving into a treasure trove.
weakness
The weakness of the project is somewhat linked to its strength because the decision was reached on creating images out of the data given at a fairly early stage. I feel that at times there was greater focus on the resulting images with far less focus on elegant code solutions. This made for made a trial and error sort of process which as a learning experience was great but slow at times. Deciding on images at an early stage I think also hindered the exploration of other possibilities for outcomes.
**Through my investigation of the precedents about the practice of using text for its formality rather than for its information carrying properties I came upon a movement and a project which informed this project quite substantially during its process. One being the French movement Letrisme which explored the the possibilities of image creation with letters, by concentrating on their formal character rather than words. And in a more contemporary project is Leonel Moura’s ISU in which there is a robot that randomly writes letters on a given surface. In ISU Genomics the robot is even programmed to draw a body in its entirety by letters.
